One of the requirements of the new alignments code is that duplicate samples are allowed. As this code will use the existing genotype decoding machinery, we'll need to relax that requirement in the existing code.